Skip to content

Vision

VisionDataModule #

Bases: LightningDataModule, DataModule[BatchType_co]

A LightningDataModule for image datasets.

(Taken from pl_bolts which is not very well maintained.)

name class-attribute #

name: str = ''

Dataset name.

dataset_cls class-attribute #

dataset_cls: type[VisionDataset]

Dataset class to use.

dims class-attribute #

dims: tuple[C, H, W]

A tuple describing the shape of the data.

__init__ #

__init__(
    data_dir: str | Path = DATA_DIR,
    val_split: int | float = 0.2,
    num_workers: int = NUM_WORKERS,
    normalize: bool = False,
    batch_size: int = 32,
    seed: int = 42,
    shuffle: bool = True,
    pin_memory: bool = True,
    drop_last: bool = False,
    train_transforms: Callable | None = None,
    val_transforms: Callable | None = None,
    test_transforms: Callable | None = None,
    **kwargs
) -> None

Parameters:

Name Type Description Default
data_dir str | Path

Where to save/load the data

DATA_DIR
val_split int | float

Percent (float) or number (int) of samples to use for the validation split

0.2
num_workers int

How many workers to use for loading data

NUM_WORKERS
normalize bool

If true applies image normalize

False
batch_size int

How many samples per batch to load

32
seed int

Random seed to be used for train/val/test splits

42
shuffle bool

If true shuffles the train data every epoch

True
pin_memory bool

If true, the data loader will copy Tensors into CUDA pinned memory before returning them

True
drop_last bool

If true drops the last incomplete batch

False
train_transforms Callable | None

transformations you can apply to train dataset

None
val_transforms Callable | None

transformations you can apply to validation dataset

None
test_transforms Callable | None

transformations you can apply to test dataset

None

prepare_data #

prepare_data() -> None

Saves files to data_dir.

default_transforms abstractmethod #

default_transforms() -> Callable

Default transform for the dataset.

train_dataloader #

train_dataloader(
    _dataloader_fn: Callable[
        Concatenate[Dataset, P], DataLoader
    ] = DataLoader,
    *args: args,
    **kwargs: kwargs
) -> DataLoader

The train dataloader.

val_dataloader #

val_dataloader(
    _dataloader_fn: Callable[
        Concatenate[Dataset, P], DataLoader
    ] = DataLoader,
    *args: args,
    **kwargs: kwargs
) -> DataLoader

The val dataloader.

test_dataloader #

test_dataloader(
    _dataloader_fn: Callable[
        Concatenate[Dataset, P], DataLoader
    ] = DataLoader,
    *args: args,
    **kwargs: kwargs
) -> DataLoader

The test dataloader.